A Countermeasure for DoS Attack on MAC Layer in Massive LoRaWAN

As a technology that provides a long-range wireless network at low energy consumption, LoRaWAN (Long Range Wide Area Network) rapidly gains attention in the Internet of Things (IoT) and shows its capability as a complementary solution in massive applications. However, LoRaWAN faces the challenge of DoS attacks. These attacks are hard to detect and counteract under the Aloha medium access control (MAC) protocol and the duty cycle limitation. In our previous work, LoRaLOFT was proposed for small LoRaWAN networks. It is a threshold-based malicious detection method implemented by a pre-trained Local Outlier Factor (LOF) model in machine learning. Focusing on massive LoRaWAN scenarios, this study investigates the DoS attack by occupying the downlink resources on the MAC layer. After migrating LoRaLOFT to massive scenarios, a countermeasure with an ACK denylist is proposed. Simulation results demonstrate the suitability of LoRaLOFT in massive LoRaWAN scenarios with its high detection performance. Moreover, the results also show that the countermeasure effectively reduces or eliminates the impact of DoS attacks in different attacking behaviors.